> There are lots of lines like the following in the patch(this if from the
> patch to arch/i386/kernel/head.S):
>
>
> - mov %dx,%ds; \
> - mov %dx,%es;
> + movl %dx,%ds; \
> + movl %dx,%es;
>
> While the patch to arch/i386/kernel/traps.c reads in part:
>
> #define get_seg_byte(seg,addr) ({ \
> register unsigned char __res; \
> -__asm__("push %%fs;mov %%ax,%%fs;movb %%fs:%2,%%al;pop %%fs" \
> +__asm__("pushl %%fs;movl %%ax,%%fs;movb %%fs:%2,%%al;popl %%fs" \
> :"=a" (__res):"0" (seg),"m" (*(addr))); \
> __res;})
>
> I might be totally off base here, but shouldn't all (most?) of these be
> "movw", not "movl" (16 bit moves, not 32 bit ones)?
>
> There are lots and lots of changes like this...
(presuming that you're talking about the Alan Modra patch)
Well, I'll leave the should bees to those more intimate with assembler.
What I can say is the patch works fine here so far.
-Mike
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu